← 返回 Skills 市场

Max QMD Search

Name: Max QMD Search
Author: sonnenberglauramarie-afk

作者 sonnenberglauramarie-afk · GitHub ↗ · v1.0.0

cross-platform ✓ 安全检测通过

264

总下载

当前安装

版本数

在 OpenClaw 中安装

/install max-qmd

功能描述

Local search/indexing CLI (BM25 + vectors + rerank) with MCP mode.

使用说明 (SKILL.md)

qmd

Use qmd to index local files and search them.

Indexing

Add collection: qmd collection add /path --name docs --mask "**/*.md"
Update index: qmd update
Status: qmd status

BM25: qmd search "query"
Vector: qmd vsearch "query"
Hybrid: qmd query "query"
Get doc: qmd get docs/path.md:10 -l 40

Notes

Embeddings/rerank use Ollama at OLLAMA_URL (default http://localhost:11434).
Index lives under ~/.cache/qmd by default.
MCP mode: qmd mcp.

安全使用建议

This skill appears to be a straightforward wrapper around the qmd CLI and is coherent with its description. Before installing: (1) review the GitHub repo (https://github.com/tobi/qmd) for any postinstall scripts or surprising behavior; (2) confirm you are comfortable letting the tool read and write an index under ~/.cache/qmd; (3) be aware it will call an Ollama endpoint (default http://localhost:11434) for embeddings/rerank — if you set OLLAMA_URL to a remote host, documents sent for embedding could leave your machine; prefer a local Ollama instance or explicitly set OLLAMA_URL. The ownerId mismatch in metadata is a packaging inconsistency worth double-checking with the publisher if provenance is important.

功能分析

Type: OpenClaw Skill Name: max-qmd Version: 1.0.0 The skill bundle provides an interface for 'qmd', a local file indexing and search tool. The installation process uses a public GitHub repository (github.com/tobi/qmd) and the commands described in SKILL.md are consistent with its stated purpose of BM25 and vector search. No indicators of data exfiltration, malicious execution, or prompt injection were found.

能力评估

ℹ Purpose & Capability

Name/description, required binary (qmd), and runtime instructions all align: the skill is a thin wrapper around the qmd CLI. Minor metadata mismatch: the _meta.json ownerId differs from the registry Owner ID provided, which could indicate inconsistent packaging/metadata but does not change the technical behavior.

ℹ Instruction Scope

SKILL.md only tells the agent to run qmd commands, index local files, and query the index. It references the OLLAMA_URL env var (default http://localhost:11434) and the local index path (~/.cache/qmd). These are within the expected scope for a local search tool, but the SKILL.md references an env var that is not declared in the registry metadata (requires.env), so the agent may read that variable at runtime even though it wasn't listed.

ℹ Install Mechanism

Install spec uses a Node package pointing at a GitHub repository (https://github.com/tobi/qmd) which will create the qmd binary. Installing a package directly from a GitHub repo is a common choice but carries moderate risk because postinstall scripts or repository content can run arbitrary code during install. This is not unusual for a CLI but is worth auditing the repository before installing.

ℹ Credentials

The skill declares no required environment variables, but the instructions expect OLLAMA_URL for embedding/rerank (defaulting to a localhost endpoint). No unrelated credentials are requested. The mismatch between declared env requirements and the documented OLLAMA_URL is a small inconsistency to be aware of.

✓ Persistence & Privilege

always is false and the skill does not request persistent platform privileges. The tool reads/writes its index under ~/.cache/qmd which is expected for a local indexing tool and does not modify other skills or global agent settings.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install max-qmd
安装完成后，直接呼叫该 Skill 的名称或使用 /max-qmd 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.0

Local BM25 + vector search + rerank with MCP mode

元数据

Slug max-qmd

版本 1.0.0

许可证 —

累计安装 0

当前安装数 0

历史版本数 1

常见问题

Max QMD Search 是什么？

Local search/indexing CLI (BM25 + vectors + rerank) with MCP mode. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 264 次。

如何安装 Max QMD Search？

在 OpenClaw 或 Claude Code 对话框中运行命令「/install max-qmd」即可一键安装，无需额外配置。

Max QMD Search 是免费的吗？

是的，Max QMD Search 完全免费（开源免费），可自由下载、安装和使用。

Max QMD Search 支持哪些平台？

Max QMD Search 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。

谁开发了 Max QMD Search？

由 sonnenberglauramarie-afk（@sonnenberglauramarie-afk）开发并维护，当前版本 v1.0.0。